Background

Hypertension is the primary risk factor for cardiovascular and cerebrovascular diseases globally1 and is closely linked to the development of various life-threatening and debilitating conditions, including heart disease, stroke, and kidney disease2. By 2025, it is estimated that one-third of the global population will be affected by hypertension3. As the world’s most populous country4China has witnessed a significant increase in hypertension prevalence over recent decades5. A large-scale, population-based study conducted in 2018 across 31 Chinese provinces reported that nearly 250 million individuals had hypertension, with 125 million unaware of their condition and approximately 150 million never having received antihypertensive treatment. Additionally, over 435 million (40%) Chinese adults were classified as having prehypertension6. Despite these concerning statistics, hypertension awareness and control rates have improved two- to threefold since 20027. In response to the escalating burden of hypertension, the Chinese government and public health institutions have strengthened policy efforts to enhance hypertension management8incorporating prevention and control measures into national health-related policies9.

TikTok, one of the fastest-growing social media platforms worldwide, has achieved extensive user penetration since its launch in September 201610. The platform enables users to create short videos enhanced with music, text, and special effects and became the most downloaded application globally in 2020, with 850 million total downloads11. TikTok’s user base is highly diverse, spanning multiple age groups, and its content covers a broad spectrum of topics, including health, nutrition, and beauty12,13. As an increasing number of patients seek medical information online before consulting health professionals, TikTok has emerged as a prominent social phenomenon and an expanding area of research14,15. Prior studies evaluating TikTok videos on acute pancreatitis, laryngeal cancer, liver cancer, and gallstones have demonstrated considerable variability in content quality and reliability16,17,18,19.

Despite the growing presence of health-related content on TikTok, hypertension-related videos remain largely unexamined. To assess the quality and reliability of such content on short video platforms, we conducted a comprehensive evaluation using the Global Quality Scale (GQS), the Journal of the American Medical Association (JAMA) benchmark criteria, and the modified DISCERN (mDISCERN) instrument. Our objective is to enhance public understanding of hypertension pathophysiology and treatment strategies, ultimately contributing to a reduction in its prevalence and disease burden. Additionally, we analyzed the relationships between video quality and key engagement metrics, including video duration, number of likes, comments, shares, and collections.

Materials and methods

Extract basic information

A systematic keyword search was conducted on TikTok to identify videos published before February 25, 2025. To minimize potential biases stemming from search history and personalized recommendations, data collection was performed without user account authentication. Videos were ranked according to TikTok’s default comprehensive ranking algorithm, without restrictions on publication date. Non-Chinese-language videos were excluded, resulting in an initial dataset of 160 videos.

A rigorous screening process was then applied to eliminate: (a) content thematically unrelated to the target keywords, (b) duplicate videos, and (c) commercial advertisements. Following this refinement, researchers systematically documented and analyzed eight key parameters: video titles, engagement metrics (likes, comments, collections, and shares), time since video upload day, video duration, and source classification. Video sources were categorized into four distinct groups: health professionals (subdivided into Traditional Chinese Medicine [TCM] and Western medicine practitioners), science communicators, general users, and news agencies.

Quality assessment

Two independent researchers (J. Wu & G. Wu) conducted blinded assessments using three validated evaluation tools: the GQS, the JAMA benchmark criteria, and the mDISCERN instrument20,21,22. The GQS employs a five-tier classification system (scores 1–5) to reflect increasing levels of content quality. The JAMA benchmark criteria evaluate four key dimensions—attribution, currency, content validity, and disclosure compliance—yielding composite scores ranging from 0 to 4. The mDISCERN instrument assesses five domains: clarity, relevance, traceability, robustness, and impartiality, with a maximum score of 5 (see Table S1 in the Supplementary Materials). To ensure methodological rigor, a double-blind evaluation process was implemented, preventing either researcher from accessing the other’s ratings during the initial assessment. In cases of disagreement, a third researcher was consulted to facilitate consensus.

Statistical analysis

Statistical analyses were performed using IBM SPSS Statistics for Mac, version 26.0. Continuous variables that did not satisfy the normality assumption were compared between groups using the Wilcoxon rank-sum test, while normally distributed continuous variables were analyzed with the independent samples t-test. Spearman’s correlation analysis and linear regression were employed to examine associations between variables. Continuous variables that did not follow a normal distribution are reported as median (IQR [interquartile range]), mean and standard deviation are used to describe continuous variables that are normally distributed, while categorical variables are presented as frequencies and percentages. A p-value of < 0.05 was considered indicative of statistical significance.

Results

Features of hypertension videos

A systematic search on TikTok using “hypertension” as the keyword initially identified a pool of videos, which was subsequently refined through a rigorous selection process. A total of 12 videos unrelated to the topic, 5 duplicates, and 4 advertisements were excluded, yielding a final dataset of 139 videos for analysis (Fig. 1). These videos comprised the cohort of hypertension education compliance content (Table 1).

Fig. 1
figure 1

Flow chart of video selection

Full size image
Table 1 Detailed characteristics of hypertension videos on TikTok.
Full size table

Platform engagement analysis revealed a total of 7190634 likes, 3608023 collections, 247602 comments, and 3252697 shares. Temporal distribution analysis indicated that the earliest qualifying hypertension education video was published in April 2021. As of the data collection cutoff, the median time since video upload day was 44 days (2-155). The highest engagement metrics recorded for a single video were 428051 likes, 333764 collections, 15910 comments, and 300041 shares.

Content provenance analysis classified uploaders into four primary categories. Health professionals represented the largest proportion (85.6%, n = 119), followed by general users (11.5%, n = 16), while science communicators and news agencies each accounted for 1.4% (n = 2) (Fig. 2). Among health professionals, 26.1% (31/119) were TCM practitioners, whereas 73.9% (88/119) were Western medicine practitioners.

Fig. 2
figure 2

Video source of the torus.

Full size image

Descriptive analysis of the three assessment tools revealed a consistent pattern in content quality. JAMA scores were predominantly concentrated at 2 (74.8%, 104/139), while the GQS had a mean score of 3, with the highest proportion of videos (36.7%, 51/139) receiving this rating. The mDISCERN assessment showed its highest frequency at score 2 (71.2%, 99/139), indicating a relatively uniform quality distribution across the evaluated dimensions.

Video quality assessment

Table 2 compares the impact of different uploader sources on video engagement metrics and content quality. Videos uploaded by health professionals had significantly longer durations and received more likes, collections, and shares compared to those from non-health professionals (p <0.05). Additionally, they achieved higher JAMA, GQS, and mDISCERN scores (p < 0.001) (Fig. 3).

Table 2 Comparison of video sources according to video features on tiktok.
Full size table
Fig. 3
figure 3

A violin diagram comparing three assessment tools of healthy professional and non-healthy professional.

Full size image

Correlation analysis between engagement metrics and content quality scores revealed significant associations. The JAMA score correlated with video duration, the number of likes, collections, shares, as well as the mDISCERN and GQS scores (p < 0.05). Similarly, the mDISCERN score was significantly associated with video duration, likes, collections, shares, and the GQS score (p < 0.05). Furthermore, the GQS score demonstrated correlations with video duration, likes, comments, collections, and shares (p < 0.05). Notably, the number of comments showed no significant correlation with either the JAMA score (p = 0.736) or the mDISCERN score (p = 0.494) (Fig. 4).

Fig. 4
figure 4

Correlation analysis heat map.

Full size image

Linear regression analysis

Table 3 presents the results of the linear regression analysis, demonstrating significant positive correlations between video duration and the JAMA (β = 0.235, p < 0.05), mDISCERN (β = 0.190, p < 0.05), and GQS scores (β = 0.410, p < 0.001). Additionally, the number of likes was positively associated with all three quality indicators (JAMA:

β = 0.661; mDISCERN: β = 0.500; GQS: β = 0.815, p < 0.05). Conversely, the number of comments exhibited a significant negative correlation with GQS scores (β = -0.574, p < 0.05).

Table 3 Linear regression analysis for JAMA, mDISCERN, and GQS score.
Full size table

Discussion

Summary and interpretation of key findings

This study conducted a comprehensive and systematic evaluation of hypertension-related content on TikTok, China’s largest and most widely used short-video platform23utilizing three validated assessment tools: the JAMA benchmark criteria, the GQS, and the mDISCERN instrument. The results provide valuable insights into the current state of hypertension-related video content and highlight areas for improvement to enhance public health communication.

TikTok serves as a powerful medium for disseminating hypertension-related information24. Social media platforms offer a convenient avenue for patients to access medical knowledge and seek social support25. The 139 hypertension-related videos analyzed in this study accumulated a total of 7190634 likes and were shared 3252697 times, highlighting TikTok’s potential as a health communication tool26,27. Many educational videos employ visual aids such as animation and graphics to simplify complex medical concepts. For example, 3D animations depicting vascular structural changes provide a visually intuitive approach that enhances information retention and comprehension28. However, despite TikTok’s potential, this study found that hypertension-related videos are of moderate quality and often lack depth. None of the videos achieved a perfect score across all three evaluation tools. Although previous research has shown that health professionals and organizations can effectively use social media for health promotion29the majority of the analyzed videos (119/139) were created by health professionals, yet their overall quality remained suboptimal. This could be attributed to the brevity of most videos, insufficient citation of authoritative sources, and a lack of content rigor.

Association between video quality and user engagement

To explore the relationship between video engagement and quality, we conducted correlation and linear regression analyses. Results demonstrated that video duration and the number of likes were positively associated with all three quality metrics, indicating that longer videos and those receiving higher engagement tend to provide better-quality information. Conversely, the number of comments was negatively correlated with GQS scores, suggesting that videos with more comments may receive greater scrutiny or criticism, leading to lower perceived quality ratings. Shares and collections showed no significant association with video quality metrics. The negative correlation between comments and GQS scores may be attributed to critical feedback reflecting concerns about content accuracy or clarity.

Implications for public health communication

Given the complexity and specialized nature of hypertension, educational videos on this topic should be extended in duration to ensure a more comprehensive and rigorous presentation of information18,30. Science communicators should adhere to high academic standards, cite authoritative references when discussing hypertension, and clearly convey any uncertainties in the available knowledge31. Furthermore, TikTok should strengthen its content review mechanisms for hypertension-related videos and limit the presence of advertisements32. To mitigate negative feedback in the comment section, AI-driven monitoring systems could be introduced, or professionals could actively engage in scientific discussions to guide users toward informed dialogue and curb the spread of misinformation33. The proliferation of false or unverified medical advertisements poses potential financial burdens and health risks to patients, underscoring the need for stricter content regulation34.

Study limitations and future directions

This study has several limitations. First, the three assessment tools used were originally designed for evaluating text-based content, which limits their applicability in assessing video-specific elements such as production complexity and the impact of audiovisual features. Future research could explore more tailored evaluation frameworks for short video content. Second, while we quantitatively analyzed the number of comments, we did not conduct an in-depth qualitative analysis of comment content. Understanding the nature of audience feedback—whether it reflects misinformation, user engagement, or skepticism—could provide deeper insights into how viewers perceive and interact with hypertension-related content. Lastly, while TikTok is a leading short-video platform in China, the generalizability of our findings to other social media platforms remains uncertain35. Future studies should compare content quality across different video-sharing platforms to determine whether similar trends exist and to identify best practices for effective health communication on social media.

Conclusion

This study provides a comprehensive evaluation of the quality and engagement of hypertension-related content on TikTok using three validated assessment tools. The findings indicate that while most videos were produced by health professionals, their overall quality remained moderate, primarily due to brevity, insufficient citation of sources, and a lack of content rigor. Statistical analyses revealed significant associations between video engagement metrics and content quality, highlighting the potential impact of interactive elements on health communication effectiveness. Given the complexity of hypertension, extending video duration, ensuring rigorous content validation, and providing authoritative references are essential for enhancing public health education. Additionally, platform-level improvements, such as stricter content review mechanisms and reduced dissemination of unverified advertisements, could further optimize the reliability of hypertension-related information. Future research should explore the applicability of these findings to other social media platforms and conduct in-depth analyses of audience engagement and comment content to refine digital health communication strategies.